With the announcement that rawpedia is back online, I went looking for two topics at the forefront of my things to read/learn about, flat fields Flat-Field - RawPedia and dark frames, the first one being flat fields.
Reading through the requirements to create a suitable flat field to use with your camera, is it fair to suggest that the characteristics of each lens and body varies enough that you can’t use someone else’s flat frames, ie that there is no value to an online library whereby you pick the camera and lens combination and use that, rather than create the required flat frames yourself?
I’ll admit that I haven’t read the page in full or slowly as I’m about to head out the door… but thought I’d ask the question.
I think if you want the most accurate flat field, you should make your own. However, RT comes with some and the lensfun project collects that info too, so I wouldn’t say there is no value in it, just that there is probably more value in doing your own.
Darks I get due to the stuck pixels and noise that would be sensor specific. Flats I’d have thought would be more likely to have similarity across lens and body combinations.
Flats are more compensatory for camera/lenses, whereas darks I use more for sensor thermal noise performance. Therefor, I always do darks per session (when I do them), and if I did flats, it would be to take a library of flats once and then re-use the appropriate flat for each shot. That’s combinatorial work if you have a lot of lenses and shot parameters to de-abberate. The question is if I have a certain model of camera and lens, could one produce a library of flats that could be used by other owners of similar combos. I don’t know but I’m sure Canon, Nikon, and Adobe do…
I do like the idea of taking flats with a large poster board outdoors in the sunlight… Seems like it would be easier to ensure even illumination, just not as portable as the filter idea. I wonder if it would be too bright for the more open apertures?
That depends on what you’re using the flat-fields for. You can use someone else’s flat-fields if you want to remove vignetting and they used the same camera-lens combo (lensfun’s vignetting correction parameters actually originate from someone having done that). You can’t use someone else’s flat-fields if you want to remove dust.
I frequently use my own flat-fields to remove dust when shooting panoramas… takes less than a minute to take my 1.5mm thick translucent plastic disc out and shoot a photo, and saves a lot of time afterwards.
Hi,
may I ask, if the pixel-to-pixel sensitivity variations are taken care of by the firmware in digital cameras (i.e. the manufacturer created a flatfield individually for the chip built into the given camera, which is stored in the camera and applied before the raw files are saved.) ? If so, one might use someone else’s flats (except for the dust as Morgan pointed out). But if the flatfield should correct the pixel-to-pixel sensitivity variations (as in astronomical CCD-cameras), there is no point in using a flat taken with a different camera.
Actually I did do this until recently, but now I have a new camera with a dust-free sensor and have not had a need for flat-fields yet. But when I need one then yes, I shoot one regularly, though if I shoot several panoramas in a house it should suffice to have one flat-field for the whole thing, unless I change lenses and risk having dust spots move. It’s really as simple as holding your hand up in front of the lens, but instead of a hand it’s a disc I cut out from a translucent milky-white plastic paint tub.